[OSGeo-Discuss] [Board] Request for a moratorium on AI/LLM use within OSGeo

Javier Jimenez Shaw j1 at jimenezshaw.com
Mon Nov 9 03:34:50 PST 2026


Hi Tobias

You mention a topic (how are LLMs using the OSS code), that is valid, but
it is not the topic discussed initially in this thread.
Even is talking about the usage of LLMs in the generation of our software
in OSGeo, not if we allow LLMs to be trained with our code.

And Even is not saying to impose, but to define a default:
"At least as a by-default policy. Each group could potentially reframe it,
but by default, I believe groups should be protected"

Cheers
Javier

On Sat, 7 Nov 2026 at 06:08, Tobias Wendorff via Discuss <
discuss at lists.osgeo.org> wrote:

> I understand your situation and what you've done for the FOSS community.
> To be clear, I'm not for or against a moratorium at this point. I'm
> still thinking about both sides. I just believe we should discuss this
> openly and look for practical solutions.
>
> I think we have to accept that the genie is already out of the bottle.
> LLMs have already been trained on public source code, including
> open-source projects. A moratorium will not change that, but design the
> future, how we deal with it.
>
> I also see a problem here. We publish software as free and open source
> so that others can read, use, change and build on it. Now we have a
> technology that can do this on a much larger scale. Of course, this does
> not mean that everything is legally or ethically OK. But should we
> reject the technology because of how some companies use it? Then we need
> to close up our source or development.
>
> Technology has always been driven by different interests, including
> science, business, competition and even military research. We may not
> like all of these interests, but they have also helped to create
> technologies that we all use today.
>
> I understand the concerns about copyright, fair payment and the power of
> big AI companies. These are real problems, especially for people who
> make a living from open-source software.
>
> But I think OSGeo should also look at how we can use this technology in
> a good way, support open alternatives and make sure that licenses are
> respected. But on the other hand, those license are MEANT to be open.
>
> I also believe that OSGeo shouldn't make final decisions over the heads
> of other projects (as the European Union does, for example), but rather
> provide guidelines that other projects can follow. Otherwise, OSGeo will
> quickly become a target of resentment, with people saying: "They're
> telling us how to behave".
>
>
> Am 2026-11-03 22:57, schrieb Even Rouault via Discuss:
> > So the foundation is going to approve/tolerate the use of "tools" (no
> > they aren't tools, they're privatization of human knowledge into
> > monopolies) that have stolen the work of its members, destroying their
> > business, and invalidating copyright ? What are you thinking to ? The
> > opacity of the process where the policy is discussed is also totally
> > at odds with the way an open foundation should work.
> >
> > Yes I'm vocal. Because that really deserves it.
> _______________________________________________
> Discuss mailing list
> Discuss at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/discuss
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/discuss/attachments/20261109/b307741d/attachment.htm>


More information about the Discuss mailing list